FineREX: Fine-Tuned NER-RE for Human Smuggling Knowledge Graphs

2026-06-17

AI Summary

This study addresses the challenges of automatically constructing a human smuggling knowledge graph from terminology-dense, unstructured court documents, which include poor domain adaptability, high levels of legal noise, and node duplication across long documents. To tackle these issues, the authors propose FineREX, an end-to-end knowledge graph construction framework tailored to this domain that leverages a fine-tuned large language model to jointly perform named entity recognition and relation extraction—eliminating the need for document rewriting or redundant preprocessing. Evaluated on a manually annotated dataset of 512 text segments, FineREX achieves relative improvements of 15.50% and 31.46% in entity and relation F1 scores, respectively, reduces legal noise by nearly 50%, lowers node duplication from 17.78% to 11.17%, and cuts end-to-end processing time by half.

Abstract

Court proceedings contain valuable evidence about human smuggling networks, but this information is often buried within unstructured, jargon-heavy legal documents. While large language models (LLMs) can support knowledge graph construction through automated information extraction, existing approaches rely on general-purpose models that are not tailored to the entity and relationship definitions required in this domain. We introduce FineREX, a streamlined knowledge graph construction pipeline built around a fine-tuned LLM for named entity recognition and relationship extraction (NER-RE). Using a manually annotated dataset of $512$ text chunks, FineREX achieves absolute improvements of 15.50% and 31.46% in entity and relationship F1-score, respectively, compared to a larger general-purpose baseline. These gains translate into higher-quality knowledge graphs, reducing legal noise by nearly half and lowering node duplication on long documents from 17.78% to 11.17%. By eliminating document rewriting and redundant extraction stages, FineREX also reduces end-to-end processing time by 50.0%. Our results demonstrate that domain-specific fine-tuning can substantially outperform larger general-purpose models while improving both the quality and efficiency of knowledge graph construction for illicit network analysis.

Problem

• human smuggling
• knowledge graph
• named entity recognition
• relationship extraction
• legal documents

Innovation

• fine-tuned LLM
• named entity recognition
• relationship extraction
• knowledge graph construction
• domain-specific NER-RE